A new automatic image-based system for measuring of prime/tack coating

The coating is one of the most important components that affect the efficiency of the pavements. The uniform distribution of coating has received less attention in pavement design and implementation. Therefore, in this study, an automatic system was presented based on image processing. The quality of the image is first enhanced and then compression and noise reduction are implemented. The thresholding is used to separate the coating from the background. After the thresholding, various properties such as the area of the coating, coefficient of variation, local maximum, and minimum are obtained from the images and used to evaluate and coating distribution. Finally, categorisation algorithms were selected for effective features in image categorisation. The data of the classification of images were compared using a confusion matrix. The results showed that the presented system has a precision of 86%. In addition, using the effective parameters in the model, the uniform distribution index was calculated to be between 0 and 100, indicating the best and worst distributions, respectively.